DeepSeek open-sources Super-Agent, a coding agent that watches Feishu chats and ships PRs
dl_weekly · x · 2026-11-09
DeepSeek has open-sourced dsh-superagent on GitHub under an MIT licence, an experimental coding agent aimed at team collaboration. The company calls it a research prototype, "not an official DeepSeek product"; Bloomberg reports it targets Anthropic's Claude Code.
Key design points:
- Minimal intrusion: works on a single GitHub repo and listens to Feishu group chats. When an approved member asks for work, it opens a task, handles it in its own session, and replies in the original thread — e.g. with a pull request. A web panel lists items needing human attention.
- Save human attention: the goal is to spare teams from coordinating and babysitting multiple agents. The manual puts it in one line: "Take as little human attention as possible, and give as much value as possible."
- Restrained execution: it acts only when asked or when team rules allow it, to avoid an agent acting for a whole team wasting the whole team's attention.
Caveats: the public code differs from DeepSeek's internal version, and some ideas from the team's write-up are not yet built.
More from coding & agent
- dair-ai founder: own your evals — public benchmarks won't tell you what to ship — dair_ai · 2026-11-09
- Open-source video-to-game: agent rebuilds 31 playable games from clips, ~$6 each — songguoxiansen · 2026-11-09
- 'Agent swarms' is funny branding — you wouldn't call coworkers a swarm — round · 2026-11-09
- 179 blocked tool calls were mostly false positives; a re-verification loop is coming — yb2698 · 2026-11-09
- 179 reward-hacking tool calls blocked during agent RL training, just 0.04% of total — yb2698 · 2026-11-09
- Agent RL for three models cost $1,177 total; LFM3 gained 43% tool efficiency — yb2698 · 2026-11-09